CN108897867A - For the data processing method of knowledge question, device, server and medium - Google Patents
For the data processing method of knowledge question, device, server and medium Download PDFInfo
- Publication number
- CN108897867A CN108897867A CN201810712464.3A CN201810712464A CN108897867A CN 108897867 A CN108897867 A CN 108897867A CN 201810712464 A CN201810712464 A CN 201810712464A CN 108897867 A CN108897867 A CN 108897867A
- Authority
- CN
- China
- Prior art keywords
- target
- answer
- intention
- question
- knowledge
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/03—Credit; Loans; Processing thereof
Landscapes
- Business, Economics & Management (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Engineering & Computer Science (AREA)
- Development Economics (AREA)
- Economics (AREA)
- Marketing (AREA)
- Strategic Management (AREA)
- Technology Law (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
Provided in the embodiment of the present invention it is a kind of for the data processing method of knowledge question, device, server and medium, the method includes：Intention assessment is carried out to the target query voice received, obtain target intention and current key information, based on the intention frame constructed in advance, determine whether current key information meets the query demand for target intention, if being unsatisfactory for query demand, it is then obtained by way of human-computer interaction and meets target critical information corresponding to query demand, be based on target intention and target critical information, target answer corresponding with target query voice is searched in the knowledge base pre-established.The embodiment of the present invention solves the problems, such as can not directly to hit when furnishing an answer for user the true intention of user, it is able to ascend the accuracy rate and integrality of user's intention assessment, to find answer feedback in credit card knowledge base to user, to accurately meet user demand.
Description
Technical field
The present embodiments relate to field of computer technology more particularly to a kind of data processing sides for knowledge question
Method, device, server and medium.
Background technique
Human-computer interaction is the science of the interactive relation between research system and user.System can be various machines
Device is also possible to the system and software of computerization.Various artificial intelligence systems may be implemented by human-computer interaction, for example, intelligence
Energy customer service system, speech control system etc..
In various human-computer interaction products, such as in the question answering system in credit card field, fully understand that user's is true
Sincere figure and true background information, for providing professional accurately answer and suggesting extremely important.Currently, being usually in spy
The various credit card relevant issues of user are answered on fixed platform by special expert, are then other use by way of retrieval
Family provides reference.But when not clear enough when the user the problem of, generally require all to demonstrate various problems by retrieving, make
Obtaining answer becomes very tediously long, the true intention for the user that cannot directly hit.
Summary of the invention
There is provided in the embodiment of the present invention it is a kind of for the data processing method of knowledge question, device, server and medium, with
Solve the problems, such as the user's true intention that can not directly hit when furnishing an answer for user in the prior art.
In a first aspect, providing a kind of data processing method for knowledge question, the method in the embodiment of the present invention
Including：
Intention assessment is carried out to the target query voice received, obtains target intention and current key information；
Based on the intention frame constructed in advance, determine whether the current key information meets the inquiry for target intention
Demand, wherein define intention in the intention frame and meet key message set corresponding to its query demand；
If being unsatisfactory for the query demand, is obtained and met corresponding to the query demand by way of human-computer interaction
Target critical information；
Based on the target intention and target critical information, searched and target query voice in the knowledge base pre-established
Corresponding target answer.
Second aspect additionally provides a kind of data processing equipment for knowledge question, the dress in the embodiment of the present invention
Set including：
Intention assessment module obtains target intention and works as carrying out intention assessment to the target query voice received
Preceding key message；
Key message determining module, for whether determining the current key information based on the intention frame constructed in advance
Meet be directed to target intention query demand, wherein defined in the intentions frame intention and meet its query demand it is right
The key message set answered；
Key message obtains module and passes through people if being unsatisfactory for the query demand for the current key information
The mode of machine interaction, which obtains, meets target critical information corresponding to the query demand；
Answer searching module, for being based on the target intention and target critical information, in the knowledge base pre-established
Search target answer corresponding with target query voice.
The third aspect additionally provides a kind of server in the embodiment of the present invention, and the server includes：
One or more processors；
Storage device, for storing one or more programs,
When one or more of programs are executed by one or more of processors, so that one or more of processing
Device realizes the data processing method for knowledge question as described in any embodiment of the present invention.
Fourth aspect additionally provides a kind of computer readable storage medium in the embodiment of the present invention, is stored thereon with calculating
Machine program realizes the data processing for knowledge question as described in any embodiment of the present invention when the program is executed by processor
Method.
A kind of data processing method for knowledge question is provided in the embodiment of the present invention, to the target query received
Voice carries out intention assessment, obtains target intention and current key information, based on the intention frame constructed in advance, works as described in determination
Whether preceding key message meets the query demand for target intention, wherein defines intention and satisfaction in the intention frame
Key message set corresponding to its query demand is obtained by way of human-computer interaction if being unsatisfactory for the query demand
It takes and meets target critical information corresponding to the query demand, the target intention and target critical information are based on, preparatory
Target answer corresponding with target query voice is searched in the knowledge base of foundation.The embodiment of the present invention solve in the prior art for
User can not directly hit when furnishing an answer the true intention of user the problem of, be able to ascend user's intention assessment accuracy rate and
Integrality, to find answer feedback in credit card knowledge base to user, to accurately meet user demand.
Detailed description of the invention
Fig. 1 is the flow chart of the data processing method for knowledge question provided in the embodiment of the present invention one；
Fig. 2 is the flow chart of the data processing method for knowledge question provided in the embodiment of the present invention two；
Fig. 3 is the structural schematic diagram of the data processing equipment for knowledge question provided in the embodiment of the present invention three；
Fig. 4 is a kind of structural schematic diagram of the server provided in the embodiment of the present invention four.
Specific embodiment
The present invention is described in further detail with reference to the accompanying drawings and examples.It is understood that this place is retouched
The specific embodiment stated is used only for explaining the present invention rather than limiting the invention.It also should be noted that in order to just
Only the parts related to the present invention are shown in description, attached drawing rather than entire infrastructure.
Embodiment one
Fig. 1 is the flow chart of the data processing method for knowledge question provided in the embodiment of the present invention one, this implementation
Example is applicable in the scene of the knowledge question of human-computer interaction be the case where problem finds answer, for example, being directed to credit card field
Human-computer interaction question and answer scene.This method can be executed by the data processing equipment for knowledge question, which can adopt
It is realized, and can be integrated on any server with network communicating function with the mode of software and/or hardware.
As shown in Figure 1, the data processing method for being used for knowledge question may include：
S101, intention assessment is carried out to the target query voice received, obtains target intention and current key information.
Various artificial intelligence systems may be implemented by human-computer interaction, for example, intelligent customer service system, speech control system etc.
Deng.User can use the man-machine interactive platforms such as similar intelligent customer service system, speech control system and put question to, when user proposes
After problem, man-machine interactive platform can provide corresponding answer by server.In other words, user is defeated to man-machine interactive platform
Enter corresponding " problem ", " problem " that then man-machine interactive platform can be inputted according to user exports corresponding " answer." example
Property, by taking the question and answer in credit card field as an example, due to current market Shang Ge big bank and local Small-Medium Banks be proposed it is more
Up to thousands of kinds of credit card, and the characteristics of each credit card has oneself, to cause user when selecting credit card without institute
It follows, just needs the true intention and true background information by fully understanding user at this time, provide professional standard for user
True answer and suggestion.
In embodiments of the present invention, target query voice can be understood as " problem " of user's input, be used for knowledge question
Data processing equipment can receive the target query voice of user's input, and the target query voice received is intended to
Identification.By carrying out intention assessment to the target query voice received, can be dug from the target query voice that user inputs
Excavate the target intention and current key information of user.Target intention can be understood as using in the target query voice that user inputs
The true intention at family.Illustratively, for credit card field question and answer can there are many users to be intended to, such as " Shen Kaliu
Journey, " " card type selection, " " Shen strip part " etc., but may not include above-mentioned complete in the target query voice of user's input
Portion is intended to, and needs to carry out intention assessment processing to the target query voice that user inputs thus, determines the true target of user
It is intended to.
In embodiments of the present invention, if it is desired to be exported accurately " answer " for the target intention of user, it is also necessary to from
Corresponding key message is excavated in the target query voice of user's input, therefore is carried out to the target query voice received
When intention assessment obtains target intention, it can also excavate to obtain the current key information for including in target query voice.It is exemplary
, by taking the question and answer in credit card field as an example, if the target meaning of the expressed user in the target query voice of user's input
Figure is " Shen strip part ", between not had nothing in common with each other by the Shen strip part of the credit card in card for different bank, in order to user
It exporting accurately " answer ", it is necessary to instruction manual needs to inquire the relevant informations such as any credit card of which bank of family, and
Often in the unobstructed target query voice that may be embodied in user's input of these information, need to excavate from target query voice
Analysis can just obtain corresponding current key information.
In embodiments of the present invention, the data processing equipment when the target query voice received, for knowledge question
Target query voice can be converted to corresponding target query text information, target is looked into using preset semantic analysis model
It askes text information and carries out speech analysis, from the target intention and target query voice for obtaining user in target query text information
The current key information for including.
S102, based on the intention frame constructed in advance, determine whether current key information meets looking into for target intention
Inquiry demand, wherein be intended to define intention in frame and meet key message set corresponding to its query demand.
In embodiments of the present invention, after target intention and current key information obtained in the target query voice, when
Preceding key message can be able to satisfy the query demand for target intention, it is also possible to be unsatisfactory for the query demand for target intention.
Illustratively, it is assumed that the target intention for including in the target query voice of user's input is " Shen strip part ", current key information
For " today applies for credit card ", current key information is unsatisfactory for the query demand for target intention at this time, in other words, according to
" today applies for credit card " these information can not make answer for " Shen strip part "；Assuming that the target query voice of user's input
In include target intention be " Shen strip part ", current key information is " XXX bank YYY class credit card ", at this time current key
Information can meet the query demand for target intention, in other words, according to " XXX bank YYY class credit card " these information
Answer can be made for " Shen strip part " to user.Based on the above situation, it is obtained from target query voice due to not knowing
Current key information whether meet the query demand breath for target intention, thus can be based on the intention frame constructed in advance
Frame, determines whether current key information meets the query demand for target intention.Wherein, it is intended that defined in frame intention and
Meet key message set corresponding to its query demand.By being intended to defined in the intention frame that constructs in advance and meeting it
Key message set corresponding to query demand may determine that whether current key information meets the query demand of target intention.
If S103, being unsatisfactory for query demand, is obtained and met corresponding to query demand by way of human-computer interaction
Target critical information.
In embodiments of the present invention, if it is determined that current key information is unable to satisfy the query demand for target intention,
It then initiates to inquire to user by way of human-computer interaction, actively be putd question to by multiple human-computer interaction to user, inquiry and target
It is intended to the corresponding key message that can satisfy query demand, as target critical information, is determined and used by multiple online interaction
Target intention and target critical information in the true intention at family, are mentioned by way of the acquisition target critical information of human-computer interaction
Rise the accuracy rate and integrality that intention assessment is carried out to the target query voice received.If it is determined that current key information can be with
Meet the query demand for being directed to target intention, then will carry out the current pass that intention assessment obtains to the target query voice received
Key information no longer passes through human-computer interaction as target critical information corresponding to the query demand for target intention is met at this time
Mode user is putd question to.
S104, it is based on target intention and target critical information, is searched and target query language in the knowledge base pre-established
The corresponding target answer of sound.
In embodiments of the present invention, it can store the knowledge question data knot pre-established in the knowledge base pre-established
Structure, can store in knowledge question data structure at least one be intended to, corresponding answer and for required for intent query
Key message.Wherein, it is each be intended to, corresponding answer and between key message required for intent query exist reflect
Penetrate relationship.It, can be by target intention and target critical information input to pre- after determining target intention and target critical information
In the knowledge base first established, match in the knowledge library lookup that pre-establishes and target intention and target critical information at least one
A corresponding answer, as target answer corresponding with target query voice.That is, pre-established in the embodiment of the present invention
Knowledge base can be understood as answering user by the answer set of input the proposed problem of target query voice, clear
In the case where user's true intention, the authoritative answers of its problem are provided.It is answered finding target corresponding with target query voice
Case can feed back to user by way of voice or text, specifically how feed back no longer specific here limit.
In addition, by taking the knowledge question in credit card field as an example, the data provided in an embodiment of the present invention for knowledge question
Processing method can help user to promote credit card related information more in credit card field by taking turns interactive dialogue mode
Efficiency is obtained, it is final that user is guided to bid to host credit card in related platform.Wherein, related platform can be understood as one and be integrated with
Question and answer, encyclopaedia, the social credit card knowledge question platform in one, such as Baidu's feed-bot platform.
A kind of data processing method for knowledge question is provided in the embodiment of the present invention, to the target query received
Voice carries out intention assessment, obtains target intention and current key information, based on the intention frame constructed in advance, works as described in determination
Whether preceding key message meets the query demand for target intention, wherein defines intention and satisfaction in the intention frame
Key message set corresponding to its query demand is obtained by way of human-computer interaction if being unsatisfactory for the query demand
It takes and meets target critical information corresponding to the query demand, the target intention and target critical information are based on, preparatory
Target answer corresponding with target query voice is searched in the knowledge base of foundation.The embodiment of the present invention solve in the prior art for
User can not directly hit when furnishing an answer the true intention of user the problem of, be able to ascend user's intention assessment accuracy rate and
Integrality, to find answer feedback in credit card knowledge base to user, to accurately meet user demand.
Embodiment two
Fig. 2 is the flow chart of the data processing method for knowledge question provided in the embodiment of the present invention two, this implementation
Example is further progress optimization on the basis of the above embodiments.
As shown in Fig. 2, the data processing method for being used for knowledge question may include：
S201, intention assessment is carried out to the target query voice received, obtains target intention and current key information.
It in embodiments of the present invention, can be by way of based on machine learning model or rule template using a large amount of instruction
Practice sample to be trained, construct target query voice for being inputted to user carry out intention assessment intention assessment model or
Intention assessment rule template, so that the intention assessment model or intention assessment rule template by building are realized to target query
The intention assessment of voice, to obtain target intention and current key information.Illustratively, it is with the question and answer in credit card field
Example, can be based on UNIT platform (the self-built user of Baidu be intended to understand machine learning platform), using large-scale training sample structure
The intention assessment model towards credit card question and answer has been built, intention assessment is carried out to user's input content to realize.In addition, in order to
Guarantee the accuracy of intention assessment model or intention assessment rule template, verify data pair can also be used according to preset time
Intention assessment model perhaps verified and according to verification result to intention assessment model or intention by intention assessment rule template
The parameter of recognition rule template is modified, and obtains revised intention assessment model or intention assessment rule template.
S202, it is intended to extract at least one corresponding keyword of every kind of intention in sample from history.
In embodiments of the present invention, every kind of intention requires one or more keywords as parameter, for user's
Different demands are distinguish, by every kind be intended to corresponding keyword can intention to user definitely, convenient for for using
The accurate answer of content search of family input.Illustratively, equally by taking the question and answer in credit card field as an example, the history of the present embodiment
Being intended to sample can be understood as counting obtained user in history word when inquiring the relevant issues of credit card, and history is intended to
The intention extracted in sample can be understood as " Shen strip part ", and corresponding keyword can be understood as " XXX bank " and " YYY class
Credit card ", keyword in this way can make more accurate inquiry for the intention of user.By being intended to corresponding key
The auxiliary of word can promote the accuracy to user's intent query, can be intended to extract every kind of intention pair in sample thus from history
At least one keyword answered inquires intention as auxiliary by the keyword of extraction.
S203, at least one corresponding keyword of every kind of intention is excavated by the method for Semantic Clustering, is obtained every
Kind is intended to corresponding key message set, and building is intended to frame.
In embodiments of the present invention, user may use keyword corresponding with being intended to similar when puing question to
Synonym is putd question to, and so-called synonym can be understood as different sayings but be intended by the table of same part thing
Up to vocabulary.At this point, may be limited due to being intended to the corresponding keyword of every kind extracted in sample intention from history, if with
Family is putd question to using synonym corresponding with keyword, is accurately returned it would be possible that can not just make for the intention of user
It answers.It based on the above situation, can be with structure after being intended to extract at least one corresponding keyword of every kind of intention in sample from history
The term vector for building each keyword, the method that may then pass through Semantic Clustering are intended to corresponding each keyword to every kind and carry out
It excavates, is intended to corresponding more keywords so as to excavate every kind.It is corresponded to according to the more every kind of intentions excavated
Keyword can construct every kind and be intended to corresponding keyword set, every kind of intention can correspond to a keyword set at this time
It closes.After obtaining the corresponding key message set of every kind of intention, it can use every kind of intention and every kind be intended to corresponding key
Information aggregate building is intended to frame.Wherein, it is intended that in frame include each intention and meet it respectively for each intention and look into
Key message set corresponding to inquiry demand.
S204, based on the intention frame constructed in advance, determine whether current key information meets looking into for target intention
Inquiry demand, wherein be intended to define intention in frame and meet key message set corresponding to its query demand.
In embodiments of the present invention, obtaining current key information from target query voice can be able to satisfy for target intention
Query demand, it is also possible to be unsatisfactory for the query demand for target intention.By current key information and it is intended to defined in frame
The keyword met in key message set corresponding to target intention query demand is compared, if current key information with
The keyword met in key message set corresponding to target intention query demand is consistent, i.e., exists in current key information full
In key message set corresponding to foot-eye intent query demand when necessary keyword, then it is assumed that current key information is full
Query demand of the foot acupuncture therapy to target intention；If meeting target intention inquiry need defined in current key information and intention frame
Ask the keyword in corresponding key message set inconsistent, i.e., there is no meet target intention inquiry in current key information
In key message set corresponding to demand when necessary keyword, then it is assumed that current key information is unsatisfactory for anticipating for target
The query demand of figure.
Furthermore, it is intended that meeting defined in frame may packet in key message set corresponding to target intention query demand
Meet key corresponding to target intention query demand for example, target intention is " Shen strip part " containing a plurality of types of keywords
It is needed in information aggregate comprising " XXX bank " and " YYY class credit card " two kinds of keyword.At this point, by current key
Information and the keyword for being intended to meet defined in frame in key message set corresponding to target intention query demand carry out
When comparison, needs the keyword for each type that will include in current key information and meet corresponding to target intention query demand
Key message set in the keyword of each type compared one by one.If each type in current key information
Keyword lacks the keyword for meeting any one type in key message set corresponding to target intention query demand, then
Current key information is considered as to be unsatisfactory for being directed to the query demand of target intention；If each type in current key information
Keyword hits the keyword for meeting each type in key message set corresponding to target intention query demand, then
It is considered as current key information and meets the query demand for being directed to target intention.It should be noted that being looked into for meeting target intention
The keyword of each type in key message set corresponding to inquiry demand, the keyword of each type is in key message set
It is middle to have different sayings, it can be the keyword of synonym property.
Illustratively, meet the keyword of each type in key message set corresponding to target intention query demand
It can be understood as wanting keyword necessary to the real meaning of target intention in the target query voice of clear user's input.It is false
If meeting in key message set corresponding to target intention query demand for target intention comprising tri- type of A, B, C
The keyword of type, each type of keyword can be using other words replacements with keyword with same saying.It will work as
The keyword for each type for including in preceding key message and meet key message set corresponding to target intention query demand
In the keywords of A, B, C three types be compared.If in current key information also including the key of A, B, C three types
Word, then it is assumed that meet key message collection corresponding to target intention query demand defined in current key information and intention frame
Keyword in conjunction is consistent, determines that current key information meets the query demand for being directed to target intention；If current key information
In include the keyword of each type lack pass any in the keyword of A, B, C three types, such as lacking type-A
Keyword, then it is assumed that meet key message corresponding to target intention query demand defined in current key information and intention frame
Keyword in set is inconsistent, determines that current key information is unsatisfactory for the query demand for target intention.
If S205, being unsatisfactory for query demand, is obtained and met corresponding to query demand by way of human-computer interaction
Target critical information.
In embodiments of the present invention, meet target if lacking in the keyword for each type for including in current key information
Any type of keyword in key message set corresponding to intent query demand in necessary all types of keyword
When, for example lacking the keyword of type-A, then it is assumed that current key information is unsatisfactory for the query demand for target intention.At this point,
It can be putd question to by words art template to user, the key lacked in current key information is got by multiple human-computer interaction
Word meets target critical information corresponding to query demand to obtain, and achievees the effect that accurately identify user's intention.It is exemplary
, target intention is " Shen strip part ", meet needed in key message set corresponding to target intention query demand include
" XXX bank " and " YYY class credit card " two kinds of keyword contains only " XXX bank " in current key information, still
It does not include " YYY class credit card ", it is considered that current key information has lacked the keyword of a type.At this point it is possible to
User sends question information the which type of credit card of XXX bank " may I ask be ", and user can answer " " YYY class credit
Card " ", then by way of human-computer interaction can keyword in completion current key information, and will be current after completion
Key message is as meeting target critical information corresponding to query demand.Certainly, for the data processing equipment of knowledge question
After sending question information to user, optional keyword, selection for reference can also be provided a user.In addition, current
It may be not only to lack a type of keyword in key message, a plurality of types of keywords may be lacked, needed at this time
To pass through multiple human-computer interaction, user is repeatedly putd question to, it is just available to meet target corresponding to the query demand
Key message.
S206, it is based on target intention and target critical information, is searched and target query language in the knowledge base pre-established
The corresponding target answer of sound.
In embodiments of the present invention, in knowledge base include history question and answer to and knowledge metadata, correspondingly, the structure of knowledge base
Building operation may include：The first building mode extracts the intention of each problem of history question and answer centering, which is mapped as answering
Case classification；Keyword set is extracted from each problem of history question and answer pair according to the key message dictionary pre-established；Root
According to answer classification and keyword set, to history question and answer to and knowledge metadata collect, obtain at least one related question and answer
To set.And/or second of building mode, it maps and is obtained belonging to each knowledge metadata from the title of knowledge metadata
Answer classification and keyword set；According to answer classification and keyword set, to history question and answer to and knowledge metadata return
Collection obtains at least one related question and answer to set.Wherein, related question and answer are to may include question and answer pair and/or Knowledge Element in set
Data, and related question and answer are to the question and answer pair and/or knowledge metadata answer classification having the same and keyword set in set.
In embodiments of the present invention, by taking the construction of knowledge base in credit card field as an example, it is based on major credit card hair fastener platform
Mass data, the data source of knowledge base is mainly all kinds of credit card metadata and history question and answer to data etc..By above-mentioned
The knowledge base of the combination building of two kinds of building modes can all fully utilize these two types of data, improve reliable answer
Recall rate.The embodiment of the present invention defines " answer classification, " " keyword set " and " related question and answer to " these three field structures,
To be constructed to knowledge base.Credit card metadata and history question and answer are elaborated separately below to the construction of knowledge base side of data
Method.
(1) for the first building mode, the data source of knowledge base is history question and answer to data.
Answer classification：The intention for extracting each problem of history question and answer centering, is mapped as corresponding answer for the intention of extraction
Classification.Specifically, can be by carrying out intention assessment to each problem of history question and answer centering, according to the intention pair of each problem
The answer of history question and answer centering is classified, and intention is mapped as answer classification, so that it is determined that each is intended to corresponding answer
Classification.
Keyword set：Pass is extracted from each problem of history question and answer pair according to the key message dictionary pre-established
Keyword set.User is intended to define specifically, each intention requires one or more key messages as parameter.For
This, can provide the dictionary for safeguarding the key message of each type, such as《Bank Name dictionary》, according to the pass of each type
The dictionary of key information matches from each problem of history question and answer pair and extracts keyword set.
Related question and answer pair：According to answer classification and keyword set, to history question and answer to and knowledge metadata collect,
At least one related question and answer is obtained to set.Specifically, can be gone through all after obtaining answer classification and keyword set
History question and answer are to collecting, so that belonging to each history question and answer of the same set to answer classification having the same and key
Set of words, thus available related question and answer pair.
(2) for second of building mode, the data source of knowledge base is credit card metadata.
Answer classification：Answer classification can be directly obtained by the title of metadata and the mapping relations of intention classification.It lifts
Example for, in the metadata of " China Merchants Bank YOUNG card platinum card " comprising " annual fee policy " this, and " annual fee policy " itself
And the intention that a kind of user is concerned about very much, answer classification can be mapped directly into.
Keyword set：The title of metadata can directly can be mapped directly as keyword set, without excavating.
Related question and answer pair：For the answer in credit card metadata source, since answer itself is its attribute value, have unique
Property, therefore relevant issues are sky, and answer also only has unique one.When customer problem is resolvable to be intended to and key message set
Later, it if the answer of metadata source can be hit, then can directly take out, without further calculating the similarity with problem.
It should be noted that can choose the first building mode for the construction of knowledge base of knowledge question, can also select
The combination for selecting two kinds of building modes, when selecting two kinds of building modes to combine, second of building mode needs is known according to actual
Data required for question and answer field are known to be constructed, and are no longer specifically limited here.For example, when for the knowledge in certain specific field
When question and answer, for example the construction of knowledge base in credit card field can be with building mode in the first building mode of simultaneous selection and second
In conjunction with wherein second of building mode has selected credit card metadata to be constructed.
On the basis of the above embodiments, optionally, it is based on target intention and target critical information, is known what is pre-established
Know and search target answer corresponding with target query voice in library, may include：
Based on target intention and target critical information, determine corresponding target correlation question and answer to set in knowledge base；From
Target correlation question and answer are in set, according to text similarity measurement algorithm, by the similar of text information corresponding with target query voice
The problem of degree meets preset threshold, corresponding answer was as the target answer；If there is no meeting asking for the preset threshold
Topic, then using the target correlation question and answer to the knowledge metadata in set as the target answer.
Specifically, target is answered when searching target answer corresponding with target query voice in the knowledge base pre-established
Case may more than one.At this point it is possible to preferentially in related question and answer to finding answer in the question and answer pair in set.If related question and answer
Answer to the question and answer in set to there is the condition that meets, can be using the answer found in question and answer pair as final mesh
Mark answer；If answer of the related question and answer to the question and answer in set to the condition that meets is not present, can be in related question and answer to set
Answer is found in middle knowledge metadata.Illustratively, when related question and answer are to answer is found in the question and answer pair in set, Ke Yitong
The method that text similarity compares is crossed to determine whether there is the target query of the problem of a question and answer pair and active user's input
The similarity of voice reaches preset threshold, if it is present, answer the problem of will just answer questions is as final target answer.
A kind of data processing method for knowledge question is provided in the embodiment of the present invention, to the target query received
Voice carries out intention assessment, obtains target intention and current key information, is intended to extract every kind of intention correspondence in sample from history
At least one keyword, at least one corresponding keyword is intended to every kind by the method for Semantic Clustering and is excavated, is obtained
It is intended to corresponding key message set to every kind, and constructs intention frame, based on the intention frame constructed in advance, works as described in determination
Whether preceding key message meets the query demand for target intention, wherein defines intention and satisfaction in the intention frame
Key message set corresponding to its query demand is obtained by way of human-computer interaction if being unsatisfactory for the query demand
It takes and meets target critical information corresponding to the query demand, the target intention and target critical information are based on, preparatory
Target answer corresponding with target query voice is searched in the knowledge base of foundation.The embodiment of the present invention solve in the prior art for
User can not directly hit when furnishing an answer the true intention of user the problem of, can not only using model judge current key letter
Whether breath meets the requirements, and can promote use by taking turns clear method online by the building of intention frame, realization more
The accuracy rate and integrality of family intention assessment, to find answer feedback in credit card knowledge base to user, thus accurately
Meet user demand.
Embodiment three
Fig. 3 is the structural schematic diagram of the data processing equipment for knowledge question provided in the embodiment of the present invention three, this
Embodiment is applicable to the case where data processing for knowledge question, which can be by the way of software and/or hardware
It realizes, and is configured on any server with network communicating function.
As shown in figure 3, the data processing equipment for being used for knowledge question may include：Intention assessment module 301, crucial letter
Determining module 302, key message acquisition module 303 and answer searching module 304 are ceased, wherein：
Intention assessment module 301, for carrying out intention assessment to the target query voice that receives, obtain target intention and
Current key information；
Key message determining module 302, for determining that the current key information is based on the intention frame constructed in advance
It is no to meet the query demand for being directed to target intention, wherein to define intention in the intention frame and meet its query demand institute
Corresponding key message set；
Key message obtains module 303 and passes through if being unsatisfactory for the query demand for the current key information
The mode of human-computer interaction, which obtains, meets target critical information corresponding to the query demand；
Answer searching module 304, for being based on the target intention and target critical information, in the knowledge base pre-established
It is middle to search target answer corresponding with target query voice.
On the basis of above scheme, optionally, which can also include being intended to
Framework establishment module 305, it is intended that framework establishment module 305 can specifically include：Keyword abstraction unit and keyword excavate single
Member, wherein：
Keyword abstraction unit extracts at least one corresponding keyword of every kind of intention for being intended to from history in sample；
Keyword excavates unit, for by the method for Semantic Clustering to every kind be intended at least one corresponding keyword into
Row excavates, and obtains every kind and is intended to corresponding key message set.
On the basis of above scheme, optionally, in knowledge base may include history question and answer to and knowledge metadata；Accordingly
, which can also include：Construction of knowledge base module 306, wherein construction of knowledge base mould
Block 306 can specifically include：First extraction unit, the second extraction unit, map unit and unit is collected, wherein：
The intention is mapped as answer class for extracting the intention of each problem of history question and answer centering by the first extraction unit
Not；
Second extraction unit, for being mentioned from each problem of history question and answer pair according to the key message dictionary pre-established
Take out keyword set；
Map unit obtains answer classification belonging to each knowledge metadata for mapping from the title of knowledge metadata
And keyword set；
Collect unit, for according to the answer classification and keyword set, to history question and answer to and knowledge metadata into
Row collects, and obtains at least one related question and answer to set, wherein the correlation question and answer are to the question and answer pair and/or knowledge in set
Metadata answer classification having the same and keyword set.
On the basis of above scheme, optionally, answer searching module 304 may include：Target correlation question and answer are to set
Determination unit, first object answer determination unit and the second target answer determination unit, wherein：
Target correlation question and answer are to set determination unit, for being based on the target intention and target critical information, described
Determine corresponding target correlation question and answer to set in knowledge base；
First object answer determination unit, for, in set, being calculated from the target correlation question and answer according to text similarity
Method, the problem of similarity of text information corresponding with the target query voice is met into preset threshold corresponding answer as
The target answer；
Second target answer determination unit, the problem of if there is no the preset threshold is met, then by the target phase
Question and answer are closed to the knowledge metadata in set as the target answer.
On the basis of above scheme, optionally, key message obtains module 302 and can be also used for, if described current
Key message meets the query demand, then using the current key information as the target critical information.
Any embodiment of that present invention can be performed in the data processing equipment for knowledge question provided in the embodiment of the present invention
Provided in the data processing method for knowledge question, have the corresponding functional module of execution method and beneficial effect.
Example IV
Fig. 4 is a kind of structural schematic diagram of the server provided in the embodiment of the present invention four.Fig. 4, which is shown, to be suitable for being used in fact
The block diagram of the exemplary servers 412 of existing embodiment of the present invention.The server 412 that Fig. 4 is shown is only an example, is not answered
Any restrictions are brought to the function and use scope of the embodiment of the present invention.
As shown in figure 4, server 412 is showed in the form of generic server.The component of server 412 may include but not
It is limited to：One or more processor 416, storage device 428 connect different system components (including storage device 428 and processing
Device 416) bus 418.
Bus 418 indicates one of a few class bus structures or a variety of, including storage device bus or storage device control
Device processed, peripheral bus, graphics acceleration port, processor or total using the local of any bus structures in a variety of bus structures
Line.For example, these architectures include but is not limited to industry standard architecture (Industry Subversive
Alliance, ISA) bus, microchannel architecture (Micro Channel Architecture, MAC) bus is enhanced
Isa bus, Video Electronics Standards Association (Video Electronics Standards Association, VESA) local are total
Line and peripheral component interconnection (Peripheral Component Interconnect, PCI) bus.
Server 412 typically comprises a variety of computer system readable media.These media can be it is any being capable of bedding and clothing
The usable medium that business device 412 accesses, including volatile and non-volatile media, moveable and immovable medium.
Storage device 428 may include the computer system readable media of form of volatile memory, such as arbitrary access
Memory (Random Access Memory, RAM) 430 and/or cache memory 432.Server 412 can be further
Including other removable/nonremovable, volatile/non-volatile computer system storage mediums.Only as an example, storage system
System 434 can be used for reading and writing immovable, non-volatile magnetic media (Fig. 4 do not show, commonly referred to as " hard disk drive ").To the greatest extent
It is not shown in pipe Fig. 4, the disc driver for reading and writing to removable non-volatile magnetic disk (such as " floppy disk ") can be provided, with
And to removable anonvolatile optical disk, such as CD-ROM (Compact Disc Read-Only Memory, CD-ROM), number
Word optic disk (Digital Video Disc-Read Only Memory, DVD-ROM) or other optical mediums) read-write CD
Driver.In these cases, each driver can be connected by one or more data media interfaces with bus 418.
Storage device 428 may include at least one program product, which has one group of (for example, at least one) program module,
These program modules are configured to perform the function of various embodiments of the present invention.
Program/utility 440 with one group of (at least one) program module 442 can store in such as storage dress
It sets in 428, such program module 442 includes but is not limited to operating system, one or more application program, other program moulds
It may include the realization of network environment in block and program data, each of these examples or certain combination.Program module
442 usually execute function and/or method in embodiment described in the invention.
Server 412 can also be with one or more external equipments 414 (such as keyboard, direction terminal, display 424 etc.)
Communication, can also be enabled a user to one or more terminal interact with the server 412 communicate, and/or with make the clothes
Any terminal (such as network interface card, modem etc.) that business device 412 can be communicated with one or more of the other computing terminal
Communication.This communication can be carried out by input/output (I/O) interface 422.Also, server 412 can also be suitable by network
Orchestration 420 and one or more network (such as local area network (Local Area Network, LAN), wide area network (Wide Area
Network, WAN) and/or public network, such as internet) communication.As shown in figure 4, network adapter 420 passes through bus 418
It is communicated with other modules of server 412.It should be understood that although not shown in the drawings, can be used in conjunction with server 412 other hard
Part and/or software module, including but not limited to：Microcode, terminal driver, redundant processor, external disk drive array, magnetic
Disk array (Redundant Arrays of Independent Disks, RAID) system, tape drive and data backup
Storage system etc..
The program that processor 416 is stored in storage device 428 by operation, thereby executing various function application and number
According to processing, such as realize the data processing method that knowledge question is used for provided in any embodiment of that present invention, this method can
To include：
Intention assessment is carried out to the target query voice received, obtains target intention and current key information；
Based on the intention frame constructed in advance, determine whether the current key information meets the inquiry for target intention
Demand, wherein define intention in the intention frame and meet key message set corresponding to its query demand；
If being unsatisfactory for the query demand, is obtained and met corresponding to the query demand by way of human-computer interaction
Target critical information；
Based on the target intention and target critical information, searched and target query voice in the knowledge base pre-established
Corresponding target answer.
Embodiment five
A kind of computer readable storage medium is additionally provided in the embodiment of the present invention five, is stored thereon with computer program,
The data processing side that knowledge question is used for as provided in any embodiment of that present invention is realized when the program is executed by processor
Method, this method may include：
Intention assessment is carried out to the target query voice received, obtains target intention and current key information；
Based on the intention frame constructed in advance, determine whether the current key information meets the inquiry for target intention
Demand, wherein define intention in the intention frame and meet key message set corresponding to its query demand；
If being unsatisfactory for the query demand, is obtained and met corresponding to the query demand by way of human-computer interaction
Target critical information；
Based on the target intention and target critical information, searched and target query voice in the knowledge base pre-established
Corresponding target answer.
The computer storage medium of the embodiment of the present invention, can be using any of one or more computer-readable media
Combination.Computer-readable medium can be computer-readable signal media or computer readable storage medium.It is computer-readable
Storage medium for example may be-but not limited to-the system of electricity, magnetic, optical, electromagnetic, infrared ray or semiconductor, device or
Device, or any above combination.The more specific example (non exhaustive list) of computer readable storage medium includes：Tool
There are electrical connection, the portable computer diskette, hard disk, random access memory (RAM), read-only memory of one or more conducting wires
(ROM), erasable programmable read only memory (EPROM or flash memory), optical fiber, portable compact disc read-only memory (CD-
ROM), light storage device, magnetic memory device or above-mentioned any appropriate combination.In this document, computer-readable storage
Medium can be any tangible medium for including or store program, which can be commanded execution system, device or device
Using or it is in connection.
Computer-readable signal media may include in a base band or as carrier wave a part propagate data-signal,
Wherein carry computer-readable program code.The data-signal of this propagation can take various forms, including but unlimited
In electromagnetic signal, optical signal or above-mentioned any appropriate combination.Computer-readable signal media can also be that computer can
Any computer-readable medium other than storage medium is read, which can send, propagates or transmit and be used for
By the use of instruction execution system, device or device or program in connection.
The program code for including on computer-readable medium can transmit with any suitable medium, including --- but it is unlimited
In wireless, electric wire, optical cable, RF etc. or above-mentioned any appropriate combination.
The computer for executing operation of the present invention can be write with one or more programming languages or combinations thereof
Program code, described program design language include object oriented program language-such as Java, Smalltalk, C++,
It further include conventional procedural programming language-such as " C " language or similar programming language.Program code can be with
It fully executes, partly execute on the user computer on the user computer, being executed as an independent software package, portion
Divide and partially executes or executed on remote computer or terminal completely on the remote computer on the user computer.It is relating to
And in the situation of remote computer, remote computer can pass through the network of any kind --- including local area network (LAN) or extensively
Domain net (WAN)-be connected to subscriber computer, or, it may be connected to outer computer (such as provided using Internet service
Quotient is connected by internet.)
Note that the above is only a better embodiment of the present invention and the applied technical principle.It will be appreciated by those skilled in the art that
The invention is not limited to the specific embodiments described herein, be able to carry out for a person skilled in the art it is various it is apparent variation,
It readjusts and substitutes without departing from protection scope of the present invention.Therefore, although being carried out by above embodiments to the present invention
It is described in further detail, but the present invention is not limited to the above embodiments only, without departing from the inventive concept, also
It may include more other equivalent embodiments, and the scope of the invention is determined by the scope of the appended claims.
Claims (12)
1. a kind of data processing method for knowledge question, which is characterized in that the method includes：
Intention assessment is carried out to the target query voice received, obtains target intention and current key information；
Based on the intention frame constructed in advance, determine whether the current key information meets the inquiry need for target intention
It asks, wherein define intention in the intention frame and meet key message set corresponding to its query demand；
If being unsatisfactory for the query demand, is obtained by way of human-computer interaction and meet mesh corresponding to the query demand
Mark key message；
Based on the target intention and target critical information, searched in the knowledge base pre-established corresponding with target query voice
Target answer.
2. the method according to claim 1, wherein the building operation for being intended to frame includes：
It is intended to extract at least one corresponding keyword of every kind of intention in sample from history；
It is intended at least one corresponding keyword to every kind by the method for Semantic Clustering to excavate, obtains every kind of intention and correspond to
Key message set.
3. the method according to claim 1, wherein in the knowledge base include history question and answer to and Knowledge Element number
According to；
Correspondingly, the building operation of the knowledge base includes：
The intention for extracting each problem of history question and answer centering, is mapped as answer classification for the intention；
Keyword set is extracted from each problem of history question and answer pair according to the key message dictionary pre-established；
Mapping obtains answer classification and keyword set belonging to each knowledge metadata from the title of knowledge metadata；
According to the answer classification and keyword set, to history question and answer to and knowledge metadata collect, obtain at least one
A correlation question and answer are to set, wherein the correlation question and answer in set question and answer pair and/or knowledge metadata is having the same answers
Case classification and keyword set.
4. according to the method described in claim 3, it is characterized in that, it is described be based on the target intention and target critical information,
Target answer corresponding with target query voice is searched in the knowledge base pre-established, including：
Based on the target intention and target critical information, determine corresponding target correlation question and answer to collection in the knowledge base
It closes；
It, will be corresponding with the target query voice according to text similarity measurement algorithm from the target correlation question and answer in set
The problem of similarity of text information meets preset threshold corresponding answer is as the target answer；
The problem of if there is no the preset threshold is met, then by the target correlation question and answer to the knowledge metadata in set
As the target answer.
5. the method according to claim 1, wherein the method also includes：
If meeting the query demand, using the current key information as the target critical information.
6. a kind of data processing equipment for knowledge question, which is characterized in that described device includes：
Intention assessment module obtains target intention and current pass for carrying out intention assessment to the target query voice received
Key information；
Key message determining module, for determining whether the current key information meets based on the intention frame constructed in advance
For the query demand of target intention, wherein define intention in the intention frame and meet corresponding to its query demand
Key message set；
Key message obtains module and passes through man-machine friendship if being unsatisfactory for the query demand for the current key information
Mutual mode, which obtains, meets target critical information corresponding to the query demand；
Answer searching module is searched in the knowledge base pre-established for being based on the target intention and target critical information
Target answer corresponding with target query voice.
7. device according to claim 6, which is characterized in that described device further includes intention framework establishment module, including：
Keyword abstraction unit extracts at least one corresponding keyword of every kind of intention for being intended to from history in sample；
Keyword excavates unit, is intended at least one corresponding keyword to every kind for the method by Semantic Clustering and digs
Pick obtains every kind and is intended to corresponding key message set.
8. device according to claim 6, which is characterized in that in the knowledge base include history question and answer to and Knowledge Element number
According to；
Correspondingly, described device further includes construction of knowledge base module, including：
The intention is mapped as answer classification for extracting the intention of each problem of history question and answer centering by the first extraction unit；
Second extraction unit, for being extracted from each problem of history question and answer pair according to the key message dictionary pre-established
Keyword set；
Map unit obtains answer classification and pass belonging to each knowledge metadata for mapping from the title of knowledge metadata
Keyword set；
Collect unit, for according to the answer classification and keyword set, to history question and answer to and knowledge metadata return
Collection obtains at least one related question and answer to set, wherein the correlation question and answer are to the question and answer pair and/or Knowledge Element number in set
According to answer classification having the same and keyword set.
9. device according to claim 8, which is characterized in that the answer searching module includes：
Target correlation question and answer are to set determination unit, for being based on the target intention and target critical information, in the knowledge
Determine corresponding target correlation question and answer to set in library；
First object answer determination unit, for, in set, according to text similarity measurement algorithm, being incited somebody to action from the target correlation question and answer
Described in the problem of similarity of text information corresponding with the target query voice meets preset threshold corresponding answer is used as
Target answer；
Second target answer determination unit the problem of if there is no the preset threshold is met, then asks the target correlation
The knowledge metadata in set is answered questions as the target answer.
10. device according to claim 6, which is characterized in that the key message obtains module and is also used to, if described
Current key information meets the query demand, then using the current key information as the target critical information.
11. a kind of server, which is characterized in that including：
One or more processors；
Storage device, for storing one or more programs,
When one or more of programs are executed by one or more of processors, so that one or more of processors are real
Now such as the data processing method as claimed in any one of claims 1 to 5 for knowledge question.
12. a kind of computer readable storage medium, is stored thereon with computer program, which is characterized in that the program is by processor
Such as the data processing method as claimed in any one of claims 1 to 5 for knowledge question is realized when execution.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810712464.3A CN108897867B (en) | 2018-06-29 | 2018-06-29 | Data processing method, device, server and medium for knowledge question answering |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810712464.3A CN108897867B (en) | 2018-06-29 | 2018-06-29 | Data processing method, device, server and medium for knowledge question answering |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108897867A true CN108897867A (en) | 2018-11-27 |
| CN108897867B CN108897867B (en) | 2021-02-09 |
Family
ID=64347407
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810712464.3A Active CN108897867B (en) | 2018-06-29 | 2018-06-29 | Data processing method, device, server and medium for knowledge question answering |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108897867B (en) |
Cited By (41)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109783605A (en) * | 2018-12-14 | 2019-05-21 | 天津大学 | A kind of science service interconnection method based on Bayesian inference technology |
| CN109840276A (en) * | 2019-02-12 | 2019-06-04 | 北京健康有益科技有限公司 | Intelligent dialogue method, apparatus and storage medium based on text intention assessment |
| CN110046177A (en) * | 2019-04-24 | 2019-07-23 | 北京思特奇信息技术股份有限公司 | A kind of method, system, storage medium and computer equipment that knowledge is extracted |
| CN110232920A (en) * | 2019-06-21 | 2019-09-13 | 百度在线网络技术（北京）有限公司 | Method of speech processing and device |
| CN110321559A (en) * | 2019-06-24 | 2019-10-11 | 腾讯科技（深圳）有限公司 | Answer generation method, device and the storage medium of natural language problem |
| CN110349463A (en) * | 2019-07-10 | 2019-10-18 | 南京硅基智能科技有限公司 | A kind of reverse tutoring system and method |
| CN110442697A (en) * | 2019-08-06 | 2019-11-12 | 上海灵羚科技有限公司 | A kind of man-machine interaction method, system, computer equipment and storage medium |
| CN110442695A (en) * | 2019-02-26 | 2019-11-12 | 北京蓦然认知科技有限公司 | A kind of intelligent interaction platform training method, apparatus, equipment |
| CN110444292A (en) * | 2019-07-29 | 2019-11-12 | 北京爱医生智慧医疗科技有限公司 | Information answering method and system |
| CN110597951A (en) * | 2019-08-13 | 2019-12-20 | 平安科技（深圳）有限公司 | Text parsing method and device, computer equipment and storage medium |
| CN110784523A (en) * | 2019-10-11 | 2020-02-11 | 北京地平线机器人技术研发有限公司 | Target object information pushing method and device |
| CN111104501A (en) * | 2019-12-23 | 2020-05-05 | 中国银行股份有限公司 | Knowledge determination method and device for test service |
| CN111161068A (en) * | 2019-12-23 | 2020-05-15 | 支付宝(杭州)信息技术有限公司 | Transaction element checking method and device |
| CN111368040A (en) * | 2018-12-25 | 2020-07-03 | 马上消费金融股份有限公司 | Dialogue processing method, model training method and related equipment |
| CN111427444A (en) * | 2018-12-24 | 2020-07-17 | 中移（杭州）信息技术有限公司 | Control method and device of intelligent device |
| CN111460095A (en) * | 2020-03-17 | 2020-07-28 | 北京百度网讯科技有限公司 | Question and answer processing method and device, electronic equipment and storage medium |
| CN111522927A (en) * | 2020-04-15 | 2020-08-11 | 北京百度网讯科技有限公司 | Entity query method and device based on knowledge graph |
| CN111666006A (en) * | 2019-03-05 | 2020-09-15 | 京东方科技集团股份有限公司 | Method and device for drawing question and answer, drawing question and answer system and readable storage medium |
| CN111858876A (en) * | 2020-05-14 | 2020-10-30 | 北京捷通华声科技股份有限公司 | Knowledge base generation method and text search method and device |
| CN111858829A (en) * | 2019-04-30 | 2020-10-30 | 广东小天才科技有限公司 | Method and device for sorting multiple intents and storage medium |
| CN111881266A (en) * | 2019-07-19 | 2020-11-03 | 马上消费金融股份有限公司 | Response method and device |
| CN112035631A (en) * | 2019-12-31 | 2020-12-04 | 北京来也网络科技有限公司 | Dialogue question-answering method, device, equipment and storage medium combining RPA and AI |
| CN112035655A (en) * | 2020-09-02 | 2020-12-04 | 中国银行股份有限公司 | Bank human resource system and working method |
| CN112035647A (en) * | 2020-09-02 | 2020-12-04 | 中国平安人寿保险股份有限公司 | Question-answering method, device, equipment and medium based on man-machine interaction |
| CN112035610A (en) * | 2020-08-28 | 2020-12-04 | 康键信息技术（深圳）有限公司 | Medical field question and answer pair generation method and device, computer equipment and medium |
| CN112071436A (en) * | 2020-09-24 | 2020-12-11 | 上海依图网络科技有限公司 | Intelligent interaction method and device for epidemic situation and storage medium |
| CN112115241A (en) * | 2019-06-21 | 2020-12-22 | 百度在线网络技术（北京）有限公司 | Question answering method, device and equipment |
| CN112182177A (en) * | 2020-09-25 | 2021-01-05 | 中国建设银行股份有限公司 | User problem processing method and device, electronic equipment and storage medium |
| CN112182176A (en) * | 2020-09-25 | 2021-01-05 | 北京字节跳动网络技术有限公司 | Intelligent question answering method, device, equipment and readable storage medium |
| CN112241448A (en) * | 2019-07-18 | 2021-01-19 | 阿里巴巴集团控股有限公司 | Response information generation method, device, equipment and storage medium |
| CN112287088A (en) * | 2020-11-20 | 2021-01-29 | 四川长虹电器股份有限公司 | Intelligent man-machine interaction query method, system, computer equipment and storage medium |
| CN112395398A (en) * | 2019-08-16 | 2021-02-23 | 阿里巴巴集团控股有限公司 | Question and answer processing method, device and equipment |
| CN112394816A (en) * | 2020-11-26 | 2021-02-23 | 浙江连信科技有限公司 | Human-computer interaction based psychological service method and robot |
| CN112463949A (en) * | 2020-12-01 | 2021-03-09 | 贝壳技术有限公司 | Data recall method and system, interaction method and interaction system |
| CN112463920A (en) * | 2020-11-25 | 2021-03-09 | 联想(北京)有限公司 | Information response method and device |
| CN112527979A (en) * | 2020-11-10 | 2021-03-19 | 联想(北京)有限公司 | Information interaction method and device and readable storage medium |
| CN112597500A (en) * | 2020-12-08 | 2021-04-02 | 国汽(北京)智能网联汽车研究院有限公司 | Automobile information security risk assessment method and device, electronic equipment and storage medium |
| CN112711653A (en) * | 2019-10-24 | 2021-04-27 | 华为技术有限公司 | Man-machine interaction method and electronic equipment |
| CN112925888A (en) * | 2019-12-06 | 2021-06-08 | 上海大岂网络科技有限公司 | Method and device for training question-answer response and small sample text matching model |
| CN115086314A (en) * | 2022-06-09 | 2022-09-20 | 中国银行股份有限公司 | Interactive data processing method and related device |
| CN117151662A (en) * | 2023-08-31 | 2023-12-01 | 北京五八信息技术有限公司 | Position information processing method, device, equipment and storage medium based on AI |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2015036945A (en) * | 2013-08-15 | 2015-02-23 | 株式会社インテリジェントウェイブ | Question answering control program, question answering control server, and question answering control method |
| CN107918634A (en) * | 2017-06-27 | 2018-04-17 | 上海壹账通金融科技有限公司 | Intelligent answer method, apparatus and computer-readable recording medium |
| CN107977415A (en) * | 2017-11-22 | 2018-05-01 | 北京寻领科技有限公司 | Automatic question-answering method and device |
| CN108170859A (en) * | 2018-01-22 | 2018-06-15 | 北京百度网讯科技有限公司 | Method, apparatus, storage medium and the terminal device of speech polling |
-
2018
- 2018-06-29 CN CN201810712464.3A patent/CN108897867B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2015036945A (en) * | 2013-08-15 | 2015-02-23 | 株式会社インテリジェントウェイブ | Question answering control program, question answering control server, and question answering control method |
| CN107918634A (en) * | 2017-06-27 | 2018-04-17 | 上海壹账通金融科技有限公司 | Intelligent answer method, apparatus and computer-readable recording medium |
| CN107977415A (en) * | 2017-11-22 | 2018-05-01 | 北京寻领科技有限公司 | Automatic question-answering method and device |
| CN108170859A (en) * | 2018-01-22 | 2018-06-15 | 北京百度网讯科技有限公司 | Method, apparatus, storage medium and the terminal device of speech polling |
Non-Patent Citations (1)
| Title |
|---|
| 王肖磊: "自动问答系统的研究与应用", 《中国优秀硕士学位论文全文数据库》 * |
Cited By (59)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109783605A (en) * | 2018-12-14 | 2019-05-21 | 天津大学 | A kind of science service interconnection method based on Bayesian inference technology |
| CN111427444B (en) * | 2018-12-24 | 2022-05-10 | 中移（杭州）信息技术有限公司 | Control method and device of intelligent device |
| CN111427444A (en) * | 2018-12-24 | 2020-07-17 | 中移（杭州）信息技术有限公司 | Control method and device of intelligent device |
| CN111368040B (en) * | 2018-12-25 | 2021-01-26 | 马上消费金融股份有限公司 | Dialogue processing method, model training method and related equipment |
| CN111368040A (en) * | 2018-12-25 | 2020-07-03 | 马上消费金融股份有限公司 | Dialogue processing method, model training method and related equipment |
| CN109840276A (en) * | 2019-02-12 | 2019-06-04 | 北京健康有益科技有限公司 | Intelligent dialogue method, apparatus and storage medium based on text intention assessment |
| CN110442695A (en) * | 2019-02-26 | 2019-11-12 | 北京蓦然认知科技有限公司 | A kind of intelligent interaction platform training method, apparatus, equipment |
| CN111666006B (en) * | 2019-03-05 | 2022-01-14 | 京东方科技集团股份有限公司 | Method and device for drawing question and answer, drawing question and answer system and readable storage medium |
| CN111666006A (en) * | 2019-03-05 | 2020-09-15 | 京东方科技集团股份有限公司 | Method and device for drawing question and answer, drawing question and answer system and readable storage medium |
| CN110046177A (en) * | 2019-04-24 | 2019-07-23 | 北京思特奇信息技术股份有限公司 | A kind of method, system, storage medium and computer equipment that knowledge is extracted |
| CN111858829A (en) * | 2019-04-30 | 2020-10-30 | 广东小天才科技有限公司 | Method and device for sorting multiple intents and storage medium |
| CN112115241A (en) * | 2019-06-21 | 2020-12-22 | 百度在线网络技术（北京）有限公司 | Question answering method, device and equipment |
| CN110232920B (en) * | 2019-06-21 | 2021-11-19 | 阿波罗智联(北京)科技有限公司 | Voice processing method and device |
| CN110232920A (en) * | 2019-06-21 | 2019-09-13 | 百度在线网络技术（北京）有限公司 | Method of speech processing and device |
| CN112115241B (en) * | 2019-06-21 | 2023-09-05 | 百度在线网络技术（北京）有限公司 | Question answering method, device and equipment |
| CN110321559A (en) * | 2019-06-24 | 2019-10-11 | 腾讯科技（深圳）有限公司 | Answer generation method, device and the storage medium of natural language problem |
| CN110349463A (en) * | 2019-07-10 | 2019-10-18 | 南京硅基智能科技有限公司 | A kind of reverse tutoring system and method |
| CN112241448A (en) * | 2019-07-18 | 2021-01-19 | 阿里巴巴集团控股有限公司 | Response information generation method, device, equipment and storage medium |
| CN111881266A (en) * | 2019-07-19 | 2020-11-03 | 马上消费金融股份有限公司 | Response method and device |
| CN111881266B (en) * | 2019-07-19 | 2024-06-07 | 马上消费金融股份有限公司 | Response method and device |
| CN110444292A (en) * | 2019-07-29 | 2019-11-12 | 北京爱医生智慧医疗科技有限公司 | Information answering method and system |
| CN110444292B (en) * | 2019-07-29 | 2022-04-08 | 北京爱医生智慧医疗科技有限公司 | Information question-answering method and system |
| CN110442697A (en) * | 2019-08-06 | 2019-11-12 | 上海灵羚科技有限公司 | A kind of man-machine interaction method, system, computer equipment and storage medium |
| CN110442697B (en) * | 2019-08-06 | 2023-09-12 | 中电金信软件(上海)有限公司 | Man-machine interaction method, system, computer equipment and storage medium |
| CN110597951B (en) * | 2019-08-13 | 2024-01-05 | 平安科技（深圳）有限公司 | Text parsing method, text parsing device, computer equipment and storage medium |
| CN110597951A (en) * | 2019-08-13 | 2019-12-20 | 平安科技（深圳）有限公司 | Text parsing method and device, computer equipment and storage medium |
| CN112395398A (en) * | 2019-08-16 | 2021-02-23 | 阿里巴巴集团控股有限公司 | Question and answer processing method, device and equipment |
| CN110784523B (en) * | 2019-10-11 | 2022-08-02 | 北京地平线机器人技术研发有限公司 | Target object information pushing method and device |
| CN110784523A (en) * | 2019-10-11 | 2020-02-11 | 北京地平线机器人技术研发有限公司 | Target object information pushing method and device |
| CN112711653A (en) * | 2019-10-24 | 2021-04-27 | 华为技术有限公司 | Man-machine interaction method and electronic equipment |
| CN112925888B (en) * | 2019-12-06 | 2024-06-28 | 上海大岂网络科技有限公司 | Training method and device for question-answer response and small sample text matching model |
| CN112925888A (en) * | 2019-12-06 | 2021-06-08 | 上海大岂网络科技有限公司 | Method and device for training question-answer response and small sample text matching model |
| CN111104501A (en) * | 2019-12-23 | 2020-05-05 | 中国银行股份有限公司 | Knowledge determination method and device for test service |
| CN111161068A (en) * | 2019-12-23 | 2020-05-15 | 支付宝(杭州)信息技术有限公司 | Transaction element checking method and device |
| CN111161068B (en) * | 2019-12-23 | 2022-06-14 | 支付宝(杭州)信息技术有限公司 | Transaction element checking method and device |
| CN112035631A (en) * | 2019-12-31 | 2020-12-04 | 北京来也网络科技有限公司 | Dialogue question-answering method, device, equipment and storage medium combining RPA and AI |
| CN111460095A (en) * | 2020-03-17 | 2020-07-28 | 北京百度网讯科技有限公司 | Question and answer processing method and device, electronic equipment and storage medium |
| CN111460095B (en) * | 2020-03-17 | 2023-06-27 | 北京百度网讯科技有限公司 | Question-answering processing method and device, electronic equipment and storage medium |
| CN111522927B (en) * | 2020-04-15 | 2023-07-14 | 北京百度网讯科技有限公司 | Entity query method and device based on knowledge graph |
| CN111522927A (en) * | 2020-04-15 | 2020-08-11 | 北京百度网讯科技有限公司 | Entity query method and device based on knowledge graph |
| CN111858876A (en) * | 2020-05-14 | 2020-10-30 | 北京捷通华声科技股份有限公司 | Knowledge base generation method and text search method and device |
| CN111858876B (en) * | 2020-05-14 | 2024-05-10 | 北京捷通华声科技股份有限公司 | Knowledge base generation method, text searching method and device |
| CN112035610A (en) * | 2020-08-28 | 2020-12-04 | 康键信息技术（深圳）有限公司 | Medical field question and answer pair generation method and device, computer equipment and medium |
| CN112035647B (en) * | 2020-09-02 | 2023-11-24 | 中国平安人寿保险股份有限公司 | Question and answer method, device, equipment and medium based on man-machine interaction |
| CN112035655A (en) * | 2020-09-02 | 2020-12-04 | 中国银行股份有限公司 | Bank human resource system and working method |
| CN112035647A (en) * | 2020-09-02 | 2020-12-04 | 中国平安人寿保险股份有限公司 | Question-answering method, device, equipment and medium based on man-machine interaction |
| CN112035655B (en) * | 2020-09-02 | 2023-11-21 | 中国银行股份有限公司 | Bank manpower resource system and working method |
| CN112071436A (en) * | 2020-09-24 | 2020-12-11 | 上海依图网络科技有限公司 | Intelligent interaction method and device for epidemic situation and storage medium |
| CN112182177A (en) * | 2020-09-25 | 2021-01-05 | 中国建设银行股份有限公司 | User problem processing method and device, electronic equipment and storage medium |
| CN112182176A (en) * | 2020-09-25 | 2021-01-05 | 北京字节跳动网络技术有限公司 | Intelligent question answering method, device, equipment and readable storage medium |
| CN112527979A (en) * | 2020-11-10 | 2021-03-19 | 联想(北京)有限公司 | Information interaction method and device and readable storage medium |
| CN112287088A (en) * | 2020-11-20 | 2021-01-29 | 四川长虹电器股份有限公司 | Intelligent man-machine interaction query method, system, computer equipment and storage medium |
| CN112463920A (en) * | 2020-11-25 | 2021-03-09 | 联想(北京)有限公司 | Information response method and device |
| CN112394816A (en) * | 2020-11-26 | 2021-02-23 | 浙江连信科技有限公司 | Human-computer interaction based psychological service method and robot |
| CN112463949A (en) * | 2020-12-01 | 2021-03-09 | 贝壳技术有限公司 | Data recall method and system, interaction method and interaction system |
| CN112597500A (en) * | 2020-12-08 | 2021-04-02 | 国汽(北京)智能网联汽车研究院有限公司 | Automobile information security risk assessment method and device, electronic equipment and storage medium |
| CN115086314A (en) * | 2022-06-09 | 2022-09-20 | 中国银行股份有限公司 | Interactive data processing method and related device |
| CN115086314B (en) * | 2022-06-09 | 2024-06-21 | 中国银行股份有限公司 | Interactive data processing method and related device |
| CN117151662A (en) * | 2023-08-31 | 2023-12-01 | 北京五八信息技术有限公司 | Position information processing method, device, equipment and storage medium based on AI |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108897867B (en) | 2021-02-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108897867A (en) | For the data processing method of knowledge question, device, server and medium | |
| US11100934B2 (en) | Method and apparatus for voiceprint creation and registration | |
| US11531818B2 (en) | Device and method for machine reading comprehension question and answer | |
| CN109657054A (en) | Abstraction generating method, device, server and storage medium | |
| WO2021218028A1 (en) | Artificial intelligence-based interview content refining method, apparatus and device, and medium | |
| CN108170859A (en) | Method, apparatus, storage medium and the terminal device of speech polling | |
| CN107832439B (en) | Method, system and the terminal device of more wheel state trackings | |
| CN103678269A (en) | Information processing method and device | |
| CN108319720A (en) | Man-machine interaction method, device based on artificial intelligence and computer equipment | |
| CN108304424B (en) | Text keyword extraction method and text keyword extraction device | |
| US20220138193A1 (en) | Conversion method and systems from natural language to structured query language | |
| CN106951503A (en) | Information providing method, device, equipment and storage medium | |
| CN108491421A (en) | A kind of method, apparatus, equipment and computer storage media generating question and answer | |
| CN110287364B (en) | Voice search method, system, device and computer readable storage medium | |
| CN111651572A (en) | Multi-domain task type dialogue system, method and terminal | |
| CN113204667B (en) | Method and device for training audio annotation model and audio annotation | |
| CN104750677A (en) | Speech translation apparatus, speech translation method and speech translation program | |
| CN108920543A (en) | query and interaction method and device, computer device and storage medium | |
| CN112632244A (en) | Man-machine conversation optimization method and device, computer equipment and storage medium | |
| CN112632258A (en) | Text data processing method and device, computer equipment and storage medium | |
| TWI674517B (en) | Information interaction method and device | |
| CN111222330A (en) | Chinese event detection method and system | |
| CN110110218A (en) | A kind of Identity Association method and terminal | |
| CN117076636A (en) | Information query method, system and equipment for intelligent customer service | |
| CN113569578B (en) | User intention recognition method and device and computer equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |